As mobile handsets used for wireless communication services are becoming smaller and lighter, battery size and power is also decreasing. Consequently, the effective talk time (i.e., transmission time) of mobile computing devices, mobile phones, and the like (i.e., handsets) is reduced.
In a conventional mobile handset, the Radio Frequency (RF) power amplifier consumes most of the power consumed in contrast to the overall system of the mobile handset. Thus, the RF power amplifier having a low efficiency typically results in degradation of the efficiency for the overall system, and accordingly reduces the talk time.
For this reason, much effort has been concentrated on increasing efficiency of the RF power amplifier in the field of power amplification. In one approach, a Doherty-type power amplifier has been introduced recently as a circuit for increasing efficiency of the RF power amplifier. Unlike other conventional power amplifiers, whose efficiency is low over the low output power range, the Doherty-type power amplifier is designed to maintain an optimum efficiency over a wide output power range (e.g., in low, intermediate, and high output power ranges).
A common Doherty-type power amplifier design includes both a carrier and a peak amplifier. The carrier amplifier (i.e., power or main amplifier), which is composed of relatively small transistors, operates to maintain the optimal efficiency up to a certain low output power level. The peak amplifier (i.e., supplemental or auxiliary amplifier) operates in cooperative fashion with the carrier amplifier to maintain a high efficiency until the power amplifier, as a whole, produces a maximum output power. When the power amplifier operates within a low power output range, only the carrier amplifier is operational; the peak amplifier, being biased as a class B or C, does not operate. But, when the power amplifier operates within a high power output range, the peak amplifier is active and may introduce nonlinearity into the overall power amplifier since the peak amplifier is biased as a highly nonlinear class B or class C amplifier.
Theoretically, the above-mentioned Doherty-type power amplifier is designed to operate while meeting the linearity specification over an entire output power range and where high efficiency is maintained. However, as described above, because the Doherty-type power amplifier comprises a carrier amplifier and a peak amplifier that operate with each other, the Doherty-type power amplifier in practice does not satisfy the linearity specification (e.g., in terms of phase or gain characteristics) over the entire output power range where high efficiency is maintained.
In summary, in the above-mentioned Doherty-type power amplifier in the related art, the linearity characteristics of such a power amplification device are difficult to predict, which makes it difficult to improve such linearity characteristics because the peak amplifier is biased at a relatively constant, low DC current level, such as a current to set the peak amplifier as a class B or C amplifier.